Ion implantation involves implantation of ions of certain elements into a solid and is a standard technique used in the fabrication of semiconductor devices. The implantation of dopant atoms such as phosphorous (P), arsenic (As), and boron (B) may be used to form semiconductor junctions, while the implantation of oxygen may be used in silicon-on-oxide (SOI) devices. Additionally, it has been reported that crystalline silicon (c-Si) solar cells have been manufactured using ion implantation methods to form the p-n junctions. However, these efforts are directed towards planar crystalline silicon wafers. In current manufacturing methods for thin-film crystalline silicon solar cell (TFSC), either planar or three dimensional cells, the p-n junctions are often formed by either POCl3-based doping, or a phosphorous compound deposition or spray-on followed by annealing.
Ion implantation of P and B for forming emitters in a p-type or n-type silicon substrate followed by a suitable annealing treatment have been shown to yield solar cells with high efficiency. However, current ion implantation efforts are limited to planar, thick c-Si wafers (typically ≧200 um).
High efficiency c-Si solar cells have been made on very thin wafers, down to 45 um, by thinning down the conventional c-Si wafers from bulk silicon ingots or bricks, using integrated circuits (IC) packaging techniques. However, this approach is often not practical because of the high cost. C-Si Thin-Film Solar Cells (TFSC) may be advantageously made by depositing a thin layer of c-Si on a suitable substrate or by slicing a c-Si ingot into thin wafers using advanced wire sawing or other known techniques such as hydrogen implantation followed by annealing to cause thin wafer separation.
Often, high performance thin-film silicon substrates (TFSS) are made by depositing an epitaxial crystalline silicon layer according to chemical vapor deposition (CVD) process. Solar cells created in this epitaxial silicon deposition method may be planar or have a well defined structure. Although, in principle, any three-dimensional surface structure is possible for 3-D cells, various performance limitations make certain 3-D structures more advantageous—such as pyramidal or prism based three-dimensional surface features.
The current standard technique for the formation of selective emitters involves several steps. Usually, the full front surface of a p-type wafer is lightly doped using the POCl3 based process or a process involving spraying a phosphorous-compound followed by anneal. Then a passivating dielectric is deposited on the front surface of the silicon substrate. The regions that are desired to be metallization contacts are then selectively opened in this dielectric, usually by a laser ablation or an etch gel process. A second doping process is then carried out to selectively dope these localized regions with a high concentration of phosphorous. However, this process is often lengthy, costly, and inefficient.
When forming homogeneous emitter layers on a TFSS, controlling the dopant profile may provide higher efficiency. To maximize current collection from the solar cell, a good ‘blue response’ is required. This requires the maximum phosphorous content near the surface to be less than 1E21 cm−3 from the surface and the depth of the emitter to be preferably in the range of 0.3 to 0.5 um thick. Because of this, there is a growing need in the solar industry for shallow emitters. However, the current industrial emitter formation processes, such as POCl3-based doping, phosphosilicate (PSG) deposition, or phosphoric acid spray-on followed by in-line anneal, do not provide control of the phosphorous concentration and depth independently. Thus, the emitter characteristics are solely determined by the temperature and time used for the doping anneal.
Further, the lifetime of minority charge carriers is greatly reduced at concentrations above 1E18 cm−3. For maximum blue response this would appear to be the upper limit of the dopant concentration in the emitter. However, this would lead to very high emitter sheet resistance and high series resistance and low fill factor (FF) and low current density (Jsc). Therefore, a thin higher doped region near the surface is desired. However, current dopant profile controlling methodology is limited.